Isolation of Eleven New Spinochromes from Echinoids of the Genus

Harriet L. Newson , Duncan A. Wild , Sing Yee Yeung , Brian W. Skelton , Gavin R. Flematti , Jane E. Allan , and Matthew J. Piggott. The Journal of Or...
0 downloads 0 Views 746KB Size
NOVEMBER 1966

SPINOCHROMES FROM ECHINOIDS

Spinochrome A Monoacetatem (10) .-sPinochrome A W a treated with ketene and the acetylation products were separated on a column of acid-treated, deactivated silica gel. The Principal (27) Syntheaized by I. Singh in this laboratory.

3645

product crystallized from isooctane to give 2-hydroxy-3-acetyl-7acetoxynaphthazarin as red needles, mp 185-189' dec. Nmr spectrum in CDCIBshowed C-3 acetyl, 6 2.88; C-7 acetoxyl, 2.39. The ultraviolet spectrum gave xmX 570 mp ( e 1410), sh 525 (3150), 493 (3700), 300 (9100), sh 252 (8620); Amin 387 m p ( e 1150), 276 mp ( e 7160).

Isolation of Eleven New Spinochromes from Echinoids of the Genus Echinothrix RICHARD E. MOORE,HARJITSINOH,AND PAULJ. SCHEUER Department of Chemistry, University of Hawaii, Honolulu, Hawaii 96826 Received March 16, 1966 The Hawaiian echinoids Echinothrix diadema Linn. and E. calumaris Pallis elaborate some 30 pigments in their spines, many of them in trace amounts. Sixteen of these pigments have now been identified (the previously described echinochrome A, spinochromes A, B, C, and D, six new naphthazarin, four new juglone derivatives, and an unprecedented benzoquinone).

The literature of the pigments derived from sea urchins had long presented a bewildering array of structural proposals based on shaky evidence. Recent work' has established the existence of but six authentic compounds (one juglone and five naphthazarin derivatives). The structures of all six have now been proven by synthesis. These six compounds were isolated from a great variety of animals which had been collected in the Atlantic and Pacific Oceans as well as in the Mediterranean Sea, thereby suggesting a concise and simple solution of a complex situation. Our equanimity was jolted when we examined the calcareous portions of two Echinothrix species, viz. E. diudema Linn. and E. calamaris Pallis, from Kaneohe Bay, Oahu, and found that these echinoderms elaborate no fewer than some 30 pigments. We have identified the five known compounds (echinochrome A and spinochromesA, B, C, and D) and now wish to report the characterization of 11 new spinochromes (six naphthazarins, four juglones, and an unprecedented benzoquinone). The complex mixture of pigments was separated readily on a column of acid-treated, deactivated silica gel into eight bands which could be eluted with benzene, one with chloroform, and one with 5% methanolchloroform. The first four fractions could be further fractionated into their components by preparative thin layer chromatography; however, for the remaining, slower moving bands it was found necessary to fractionate the mixture after methylation with diazomethane and regenerate the free pigments by acid hydrolysis. The structures of the 16 identified spinochromes from Echinothrix, their Rf values, melting points, and yields are presented in Table I. Structure Determinations.-In addition to echinochrome A (12) and spinochromes A (9),B (16), C (13), and D (15), we have isolated 2-hydroxy-3-acetyl2-hydroxy-6-ethyljuglone 2naphthazarin ( (Z),394

(1) I. Singh, R. E. Moore, C. W. J. Chang, and P. J. Scheuer, J . Am. Chem. Soc., 87, 4023 (1965). (2) Methylation patterns of polyhydroxynaphthoquinones will be the subject of a forthcoming publication by C. W. J . Chang, R. E. Moore, H. Singh, and P. J. Scheuer. (3) R. E. Moore, H. Singh, C. W. J. Chang, and P. J. Scheuer, J . Ow. Chem., 82, 3638 (1966). (4) Not isolated in sufficient amount for nmr and mixture melting point with an authentic sample. This point will be clarified in a subsequent paper dealing with the structures of the remaining pigments of Echinothr+z echinoids.

hydroxy-6-ethylnaphthazarin (3), naphthopurpurin ( 5 ), 2,7-dihydroxy-6-acetyljuglone(6), 2,7-dihydroxy3-ethylnaphthazarin (8),3 and 2,7-dihydroxynaphthazarin (11)s-8 and have shown their identities by comparison with authentic samples. The most interesting pigment in this group is 2,5dihydroxy-3-ethylbenzoquinone(4), the first representative benzoquinone to have been discovered in marine invertebrates. Examination of its ultraviolet spectrum (Figure 1) showed that the pigment was not a typical naphthazarin or juglone. The structure of 4 is readily elucidated from its infrared, nmr, and mass spectra. The infrared spectrum showed a sharp band at 1613 cm-' attributed to a quinone carbonyl having an adjacent h y d r ~ x y l . ~Since this was the only carbonyl absorption, both quinone carbonyls have to be flanked by hydroxyl substituents. The molecular weight of 168 determined by mass spectrometry suggested a 2,3or 2,5-dihydroxybenzoquinone substituted with an ethyl group. The nmr spectrum immediately revealed the presence of the ethyl group and this was substantiated by the small, but nevertheless characteristic doublet at m/e 139 (34 - C2HS) and 140 (M - CH2CH2) in its mass spectrum.1° The position of the C-6 proton signal (6 5.82) is only compatible with the placement of a hydroxyl at C-5. Therefore the pigment must have structure 4. Compound 4, a previously unknown substance, was easily synthesized in 27% yield by oxidation of ethylhydroquinone with basic hydrogen peroxide. The principal pathway of disintegration in the mass spectrum of 4 begins with the loss of a methyl radical from the parent ion, the base peak, to form the second (5) 2,7-Dihydroxynaphthaaarinhas also been identified as a minor pigment in the spines of the Hawaiian echinoida Echinometra oblonga Blainville and Tripneusteo gratzlla 1 h n . (6) Under the conditione of the isolation 2,7-dihydroxynaphthazarinis not formed as an artefact from loss of the acetyl group of spinoohrome A owing to acid hydrolysis. (7) This pigment, called mompain, has recently been isolated from the microorganism Helicobasidium mompa Tanaka [S. Natori, Y. Kumada, and H. Nishikawa, Chem. Pharm. Bull. (Tokyo), 18, 633 (1965)l. After completion of our work we became aware of its syntheais by Professor R. H. Thomson (private communication) oia a tetralone intermediate [cf. H. A. Anderson, J. Smith, and R. H. Thomson, J . Chem. Soc., 2141 (196511. (8) The electronic absorption spectra and synthesis of substituted 1,4naphthoquinone8 will be presented shortly in a full paper by C. W. J. Chang, R. E. Moore, R. Ogata, I. Singh, and P. J. Scheuer. (9) 9. Natori, Chem. Pharm. Bull. (Tokyo), 18, 511 (1965). (10) The m / e 130 and m / e peaks could be due to loss of CHO radical and CO, respectively, from the molecular ion.

MOORE,SINQH,AND SCHEUER

3646

VOL.31

TABLEI PIGMENTS FROM THE SPINES OF Echinothriz d i a d a a AND E. ealamria IN ORDEROF ELUTION Relative Pigment

RfO

MP, O C

Yield, %

163-164 dec

0.0001

Pigment

Relative Rf'

MP, 'C

0.105

192-193

0.0004*

0.050

245-255 subl

0.00005*

Yield, %

OH 0 OH

0.692

COCH, OH 0

OH 0 1

9

0

0.618

219-220'

300

0.00015°

OH 0

4

12

OH0

OH 0 OH

0.412 200-210 subl

OH

0.000075

OH OH

0

13

5

0

HO

OH

0.366 215 dec

0.00004

OH 0 14

6

OH

OH 0

0

OH

0.169

CH$O

179-180 dec

OH

O.ooOo75 HO

OH

6H 0 7

15

Ho6k

0.146

OH 0

0

HO

190-192

OH

0.000356 OH

CHpCH3

OH 0 16

8

On a thin layer plate of acid-washed, deactivated silica gel with benzene; referred to naphthazarin (Rt 1.000). The Rr values are not reproducible and will vary appreciably depending on the characteristics of the plate. These Rr values are the result of a single accurate experiment and can be compared with those reported in other tables in this manuscript and all other related papers. b Insufficient amount of pigment prevented determination of melting point; reported value is that of reduction product of spinochrome A (see ref 3). 0 From spines of E. calanzaris; pigment not observed in E . d i a d e m . d From spines of E. d i a d e m ; pigment not observed in E . c a l u m r i s . e Yield based on methylated derivative. 0

most abundant ion a (m/e 153, 44% relative intensity) followed by two consecutive expulsions of carbon monoxide to give peaks at m/e 125 (b, 18% relative intensity) and m/e 97 (e, 8% relative intensity). Metastable ions" outline this course of fragmentation. metastable ion are marked by asterisks in the proposed fragmentation schemes. (11) Transitions aupported by

8

The structure of 2-hydroxy-6-acetylnaphthazarin (7) was deduced from spectral and chemical evidence. The mass spectrum12 confirmed the molecular weight of 248 and the ultraviolet spectrum (Figure 1) revealed (12) For a diacusaion of the maas spectrum of 7, see the following paper: D. Becher, C. Djerasai, R. E. Moore, € Singh, I. and P. J. Scheuer, J . 070. m m . , ai, 81x0(igstv.

NOVEMBER 1966

.1cr

& CH2

-CH3

HO

SPINOCHROMES FROM ECHINOIDS

a *

15

3647

In -I II

0

b, m/e 125

c, m/e 97

'4

a naphthazarin with one p-hydroxyl.8 The nmr specTQ trum showed the presence of an acetyl group (6 2.73) X and furthermore that both the 6-hydroxyl and acetyl substituents have adjacent ring protons. The chemical W shift of 6 6.43 for the C-3 proton and 7.63 for the C-7 proton indicated the location of the quinoid and benze5 noid properties. l 3 The 7-acetyl isomer was discarded as a possibility when 7 was converted to the known 3 by sodium borohydride reduction. Compound 7 was synthesized in 15% yield by the action of boron trifluoride on an acetic anhydride solution of 1,2,4,5,8pentaacetoxylnaphthalene (leucoacetate of naphthopurpurin) followed by mild basic hydrolysis in the presence of air.14 A small amount of 1 (2% yield) was 1 ' 1 ~ 1 1 also formed in the synthesis. 300 400 500 600 Pigment ioi5was rather difficult to handle16 but the 2,3,7-trihydroxy-6-acetyljuglonestructure was sugWAVELENQTH, mp gested as the most probable from the mass spectrum Figure 1.-Electronic absorption spectra of. 2,5dihydroxy-3(m/e 264) and qualitative electronic spectrum (A,,, ethylbenzoquinone and 2-hydroxy-6acetylnaphthazarin. 305, 388, 490 mp). The nmr spectrum of the stable dimethyl ether, 2,3-dimethoxy-6-acetyl-7-hydroxyjuglone (17)immediately confirms the proposed structure. The signals a t 6 4.10 and 4.15 indicate that both methoxyls are on the quinoid ring.la The C-8 proton appears a t 6 7.16 and thus establishes a hydroxyl a t C-7.13 Two hydroxyl signals can be seen at 6 13.95 and 14.20 showing intramolecular hydrogen bonding to proximate carbonyl functions, the C-7 hydroxyl being bonded to the acetyl group at C-6 (6 2.83).13 The ultraviolet spectrum of 17 (Figure 2) is similar to that of 6.3 Compound 17 could be hydrolyzed with concentrated hydrobromic acid to regenerate 10, but in poor yield. The structural elucidation of 2,3,7-trihydroxy-6ethyljuglone (14), the major pigment of E. diudema, offered no difficulty. The ultraviolet spectrum (Figure 2) demonstrated the juglone chromogen with three B hydroxyls.8 The presence of the ethyl group a t C-6 was revealed by the nmr spectrum and the signal a t 6 7.13 for the C-8 proton could be reconciled only \ 300 400 1 ' S1O0 ' 600 1 ' by the attachment of a hydroxyl a t C-7.I3 The pigment 300 400 SO0 600 formed a trimethyl ether and the chemical shifts for WAVELENGTH, m y the three methoxyls (see the Experimental Section) were compatible only with the assigned ~tructure.'~ Figure 2.-Electronic absorption spectra of 2,3,7-trihydroxy-0-

1

nA

1 ,U'\'

ethyljuglone and 2,3-dimethoxy-7-hydroxy-6-acetyljuglone. (13) R. E. Moore and P. J. Scheuer, J . Org. Chem., 81, 3272 (1966). (14) Synthesized by I. Singh in this laboratory. For similar experimental details 6- ref 8. (16) 2,3,7-Trihydroxy-6-acetyljugloneis also present as a minor pigment in the spines of Echinomctro oblonga Blainville. (18) Instability of 10 may be due to air oxidation or light-catalyzed reac-

tions in solution as evidenced by thin-layer chromatography and ultraviolet absorption spectrum.

The mass spectrum of 14 shows the molecular ion (m/e 250) as the base peak and its principal disintegration is initiated by expulsion of carbon monoxide to give an m/e 222 ion d (18% relative intensity) which subsequently eliminates a methyl radical to form the

VOL.31

MOORE, SINGH,AND SCHEUER

3648

m/e 207 ion e (52% relative intensity). Appropriate metastable peaks at m/e 197.1 (222a/250 = 197.1) and

193.0 (2072/222 = 193.0) accompany these fragmentations. Loss of carbon monoxide from ion e followed by rearrangement results in the m/e 179 ion f (5% relative intensity) and a metastable ion at m/e 154.8 (17g2/207 = 154.8) supports this transition.

TABLE I1 Fraction

1 2 3 4 5 6

a

Eluent

Band color

Benzene Pink Benzene Orange-yellow Benzene Pink Benzene Red Benzene Purple Benzene Orange Benzene 7 Orange-red to brown. 8 Orange-yellow Benzene 9 Chloroform Orange to red-orange 10 MethanolBrown chloroform 5:95 Fraction 7 does not separate entirely from 6.

Elution time

0 . 5 hr lhr 2-3 hr 4-6 hr 8-12 hr 1 day 1 day 2-3 days 2-3 days 2-3 hr

d, n$ 222

?H

e, m/e 207

f,

*

179

Experimental SectionI7 Isolation of Spinochromes from Echinothriz dia&ma. A. Isolation and Partial Separation of Pigment Mixture by Column Chromatography on Silica Gel.-The animals were collected periodically during June 1963-July 1965 in Kaneohe Bay, Oahu, a t a depth of 4-12 ft and the spines were removed by rubbing each of the animals on a wide mesh wire screen. Fleshy parts were separated from the calcareous portions by abrasive agitation in water and decantation of the suspended flocculent matter. After the aqueous decant had become relatively clear, the spines were washed thoroughly with acetone and dried. A 1-kg batch of spines was distributed among ten 41. beakers or flasks and each ca. 100-g portion was digested in 250 ml of concentrated hydrochloric acid. Considerable foaming occurred and the addition of acid usually required 1 hr. The mixture was filtered (filtration is slow and usually takes 3 hr) and the filtrate was extracted exhaustively with peroxide-free ether to remove the pigments. The spinochromes were washed into aqueous sodium bicarbonate solution under nitrogen and the ethereal layer was discarded. The aqueous phase which contained a considerable amount of precipitated sodium salts of the pigments was acidified with hydrochloric acid and extracted thoroughly with benzene and finally with ether. The isolation of the crude pigment required about 6-8 hr and the best yields were usually obtained if the isolation to this point was not interrupted and if the crude pigment was introduced onto the silica gel for chromatography (see below) on t.he same day. The benzene extract was evaporated and the residue was chromatographed on a 40 x 1.8 cm column of acid-treated, deactivated silica gel.' The chromatogram was developed in a continuous fashion18 with benzene, chloroform, and 5% methanolchloroform. Eight distinct bands separated with benzene elution and each band plus the intermediate zone preceding the next band constituted a fraction. Two more fractions were collected by contiripous elution with chloroform and 5% methanol(17) Analyses were performed by Berkeley Analytical Laboratory, Berkeley, Calif. In view of the notoriously poor behavior of these compounds during combustion, few samples were submitted for elemental analysis. Ultraviolet-visible spectra were determined in chloroform solution on a Cary 14 spectrophotometer, nmr spectra on a Varian A40 instrument. Mean speotra were determined at Stanford University on an A.E.I. MS-9 instrument operating with an ionieation energy of 70 ev. Melting pointa were determined on a Fisher-Johns apparatus and are unoorrected. Previously known compounds and their methyl ethers were compared by ultraviolet and nmr spectra, mixture melting points, and R, values in two systema. Diasomethane methylatiom were conducted in ether or methanol-ether solution and the progress of the reaction was followed by tlc. In many instances care had to be exercised to avoid the undesired methylation of peri hydroxyls, especially in the methylation of pigmenta possessing acetyl groups. (18) H.Ropoport and J. Bordner, J . Or#. Chsm., Sa, 2727 (1984).

chloroform. The results of the chromatography are summarized in Table 11. A total of 20 kg of E. diadema spines was processed. B. Separation of Spinochromes in Fraction 1.-The gummy red solid (fraction 1) was purified by preparative thin layer chromatography (tlc) on acid-treated, deactivated silica gel.' The plate (0.5 mm X 20 cm x 20 cm) was developed with benzene to give a rapidly moving pale yellow band followed by a red band. The pigment was recovered from the red band and crystallized from petroleum ether (bp 30-60') to give impure 2-hydroxy-3-acet ylnaphthazarin. The mass spectrum indicated the presence of a substantial amount of impurity. The pigment was methylated carefully with diazomethane and the progress of the methylation was determined on a tlc plate, on which the methylated pigment travelled slower than the free pigment. The product was purified by preparative tlc as described above. The methylated pigment was somewhat labile and hydrolyzed merely on standing for a few days in chloroform solution to give 2 mg of the natural pigment, 2-hydrory-3acetylnaphthazarin ( l ) , as dark red needles from isooctane, mp 163-164" dec. C. Separation of Spinochromes in Fraction 2.-The crude pigment in fraction 2 was rechromatographed on a column of acid-treated, deactivated silica gel and elution with benzene removed a pale yellow-brown band (fraction 2A) followed closely by an orange-yellow one (fraction 2B). Fraction 2A was homogeneous by tlc and vacuum sublimation followed by recrystallization from benzene yielded 5 mg of 2,5-dihydroxy-3-ethylbenzoquinone(4) as orange prisms, subliming a t 130-145' without melting. The ultraviolet spectrum showed Xmr sh 282 mp ( E 13,200), 288 (13,400), 422 (183); kmin 325 mp (e 118). Nmr spectrum in acetone-& showed C-3 ' a, 1.02 (tripmethylene, 62.41 (quartet, J = 7.5 cps); C-3 CH&H let, J = 7.5 cps); C-6 hydrogen, 5.82. Mass spectrum showed m/e 168 (relative intensity loo), 153 (44),140 (3), 139 (2), 125 (18), 97 (8),94(6), 69 (19). Fraction 2B was fractionated further by preparative tlc with benzene to give an orange band (fraction 2B1), a yellow band (fraction 2B2), and a yellow band (fraction 2B3). Fraction 2B1 was homogeneous by tlc and recrystallization from chloroform and vacuum sublimation gave 15 mg of n a p t h o p u r p h (5) as orange-brown crystals, subliming a t 200-210' without melting. Fraction 2B2 was also pure and crystallization from chloroform afforded 8 mg of 2,7-dihydroxy-6-acetyljuglone (6)as small, orange needles, mp 215' dec. D. Separation of Spinochromes in Fraction 4.-The crude pigment in fraction 4 was separated on preparative tlc plates with benzene into a red band (fraction 4A), an orange band (fraction 4B) which did not separate entirely from fraction 4A, and a red band (fraction 4C). Fraction 4A was homogeneous by tlc and crystallized readily from chloroform-isooctane to give 15 mg of P-hydroxyd-acetylnaphthazarin (7) as black needles, mp 179-180" dec. The ultraviolet spectrum showed XmX 236 mp ( E 16,500), 297 (7930), 507 (6450); Xmln 266 mp ( e 4980), 352 mp ( E 950). Nmr spectrum in CDCls gave C-3 hydrogen, 6 6.43; C-6 acetyl, 2.73; C-7 hydrogen, 7.63. Fraction 4C was slightly impure, but further preparative tlc did not result in satisfactory separation of the pigments. Methylation with diazomethane and preparative tlc of the methylated 4C (4CM) with benzene resulted in ready separation into a red band (fraction 4CM1), a red band (fraction 4CM2), an orange band (fraction 4CM3), and an orange-yellow band (fraction

NOVEMBER 1966

SPINOCHROMES FROM ECHINOIDS

3649

Fraction 6M1 crystallized readily from isooctane as long, red needles of 2,3,6-trimethoxy-7-ethylnaphthazarin (trimethylechinochrome A), mp 131-132" (lit.*I mp 130'). Fraction 6M2 crystallized from isooctane to give 2-hydroxy-3-ethyl-6,7dimethoxynaphthazarin as small, red needles, mp 153-154" . I 9 Fraction 6M5 crystallized from isooctane to give 2-hydroxy-3,7dimethoxy-6-ethylnaphthazarin, mp 152-154'. l 9 Fractions 6M1, 6M2, and 6M5 were combined and demethylated in concentrated hydrobromic acid (2 hr of reflux under nitrogen) to give ca. 350 mg of echinochrome A (12), as red needles from chloroform, mp 222-223'. Crystallization of fraction 6M3 from chloroform afforded 2,3,6trimethoxy-7-acetylnaphthazarin ( trimethylspinochrome C) as brown needles from isooctane-chloroform, mp 126-127" (lit.** mp 116-117'), and fraction 6M4 gave 6,7-dimethoxy-2-hydroxy3-acetylnaphthazarin, as reddish brown needles from isooctanechloroform, mp 224-227' dec. Fractions 6M3 and 6M4 were combined and hydrolyzed in 1: 1 ethanol-12 N hydrochloric acid (22 hr of reflux under nitrogen) to yield after tlc spinochrome TABLE I11 C ( l3)zaas red-orange needles from absolute methanol, mp 246Identity with other fractions 248'. Band color Fraction Crystallization of fraction 6M7 gave 2,7-dimethoxynaphthaOrange Contains some 4CM1 5AM1 zarin as dark red needles from chloroform-isooctane, mp 273-275" Yellow 5AM2 with sublimation, which after hydrolysis in concentrated hydroRed 4CM2 5AM3 bromic acid (2 hr reflux under nitrogen) resulted in 150 mg of Purple 5AM4 2,7-dihydroxynaphthazarin(1 1) as small, red needles after vacuum Red 5AM5 sublimation and crystallization from chloroform, subliming a t Yellow 5AM6 265-275" without melting (lit.' mp >300" dec). Fraction 7 was treated in exactly the same manner as fraction Red 5AM7 6 since they consisted of the same pigments, but in different Purple 5Ah18 amounts. Fraction 6 contained most of the echinochrome A 5AM9 Orange-yellow whereas fraction 7 contained most of the spinochrome C and 2,7-dihydroxynaphthazarin. G . Separation of Spinochromes in Fraction 8.-The pigment Fraction 5AM2 crystallized from isooctane to give 2,3-diin fraction 8 was essentially pure and vacuum sublimation folmethoxy-7-hydroxy-6-acetyljuglone as small, orange-yellow lowed by two recrystallizations from chloroform gave ca. 400 needles, mp 134-135'. The ultraviolet spectrum showed hmsx mg of 2,3,7-trihydroxy-6-ethyljuglone( 14) as brick-red needles, 238 mp ( e 18,000), 307 (13,300), 378 (3740), 462 (2530); Xmin mp 265-269' dec with partial sublimation. The ultraviolet 278 mp ( e 4750), 350 (2840), 423 (2030). Nmr spectrum in spectrum gave Amax 270 mp ( E 21,900), 330 (10,600), 417 (3740), CDCl3 showed C-2 methoxyl, 6 4.10; C-3 methoxyl, 4.15; 485 sh (1510); Amin 246 mp ( e 7180), 286 (1440), 367 (1970). C-5 hydroxyl, 13.95 or 14.20; C-6 acetyl, 2.83; C-7 hydroxyl, Nmr spectrum in acetone-de showed C-5 hydroxyl, 6 12.08; 13.95 or 14.20; C-8 hydrogen, 7.16. Fraction 5AM4 crystallized from chloroform-isooctane as C-6 methylene, 2.71; C-6 CHsCHa, 1.14; C-8 hydrogen, 7.13. black needles of 2-hydroxy-7-methoxy-3-acetylnaphthazarin Mass spectrum showed m/e 250 (relative intensity loo), 235 (7), (monomethylspinochrome A), mp 246-248'. Fraction 5AM5 222 (lg), 207 (52), 179 (5). Anal. Calcd for CnHloO6: C, 57.6; H, 4.0. Found: C, crystallized from isooctane to give 2,7-dimethoxy-6-acetylnaphthazarin (dimethylspinochrome A) as long, red needles, mp 57.1; H, 3.6. Methylation with diazomethane formed a trimethyl ether, 181-182'. Fraction 5AM4 and 5AM5 were combined and hydromp 113-114', as orange needles from isooctane. Nmr spectrum lyzed in 1: 1 ethanol-12 N hydrochloric acid a t reflux (nitrogen in CDC13 showed C-2 methoxyl, S 4.07; C-3 methoxyl, 4.09; atmosphere). The progress of the hydrolysis was determined by tlc and 40 mg of spinochrome A (9) was isolated after chroC-5 hydroxyl, 12.23; C-6 methylene, 2.72; C-6 CHzCH,, 1.10; C-7 methoxyl, 3.96; C-8 hydrogen, 7.20. matography.20 Fraction 5B was subjected to preparative tlc and separated into H. Separation of Spinochromes in Fraction 9.-The crude a purple band (5A), a light brick-red band (5B1), a yellow band pigment was methylated with diazomethane (9M) and chromatographed on preparative thin layer plates with benzene. (5B2) , and an orange-brown band (5B3). Fraction 5B3 crystallized from chloroform to give 3 mg of 2,3,7-trihydroxy-6-aceQl- The results of the chromatography are outlined as shown in juglone ( 10) as brick-red crystals, subliming a t 245-255' without Table V. melting. An additional amount of the pigment could be obtained Fraction 9M1 was essentially pure and crystallization from by acid hydrolysis of fraction 5AM2 ( 2 hr of reflux in concentrated isooctane and from absolute methanol yielded 2,3,6-trimethoxynaphthazarin (trimethylspinochrome D ) as red needles, mp hydrobromic acid under nitrogen), but in poor yield.lE F. Separation of Spinochromes in Fractions 6 and 7.161-162'. Hydrolysis in 1 ml of 1:1 ethanol-12 N hydrochloric acid (125" for 24 hr in sealed tube) afforded 1 mg of spinochrome Fraction 6 was carefully methylated and subjected to preparative tlc with benzene. The results of the chromatography are D (15) as brick-red crystals after vacuum sublimation, subliming summarized as shown in Table IV. at 280-290" without melting. Fraction 9M12 was a mixture and separation was better achieved by rechromatography on plates using chloroform (see TABLE IV Table VI). Fraction 9M12G crystallized from benzene &s orange crystals Fraction Band color of 2,3-dimethoxy-7-hydroxyjuglone(dimethylspinochrome B) , 6M1 Red-orange mp 206205'. 19 Hydrolysis in concentrated hydrobromic acid 6M2 Red-orange (2 hr of reflux under nitrogen) gave 15 mg of spinochrome B 6M3 Red (16)as brick-red crystals after vacuum sublimation, mp >300". 6M4 Purple 6M5 Red (20) The yield of spinochrome A is no greater than 60% (based on methylated derivative) due to hydrolytic loss of the acetyl group. Appreciable 6M6 Yellow amounts of 2-hydroxy-7-methoxynphthazarin and 2,7-dihydroxynaph6M7 Red thazarin are formed during the reaction. 6M8 Yellow (21) K. Nishibori, Nature, 184, 1234 (1959). (Remaining bands) 6M9 (22) C. W. J. Chsng, R. E. Moore, and P. J. Scheuer, Tetrahedron Letters, 4CM4). Fraction 4CM1 crystallized from isooctane as long, red needles of 2,7-dimethoxy-6-ethylnaphthazarin, mp 145-147', and fraction 4CM2 crystallized identically to give 2-hydroxy-7methoxy-3-ethylnaphthazarin, l o mp 230-232". Fractions 4CM1 and 4CM2 were combined and hydrolyzed in 1:1 ethanol-12 N hydrochloric acid (48 hr of reflux under nitrogen). Recrystallization from absolute methanol afforded 70 mg of the natural pig(8) as dark red needles, ment, 2,7-dihydroxy-3-ethylnaphthazarin .~ mp 190-192'. E. Senaration of Sninochromes in Fraction 5.-The crude pigment was rechromatographed on silica gel and partially separated into a purple band (fraction 5A) and a reddish brown band (fraction 5B). Fraction 5A was mainly spinochrome A but was somewhat impure. The pigment was carefully methylated with diazomethane (5AM) and chromatographed on tlc plates with benzene. The results are summarized in Table 111.

(19) The structural determinations of these partially methylated spinochromes will be discussed in a forthcoming publication (see ref 2).

3557 (1964). (23) The yield of spinochrome C based on the trimethyl derivative is ca. 50%. Prolonged hydrolysia (45 hr) leads to spinochrorne D in 95% yield.

BECHER, DJERASSI, MOORE,SINGH,AND SCHEUER

3650

TABLE V Fraction

Band color

9M1 9M2 9M3 9M4 9M5 9M6 9M7 9M8 9M9 9M10 9Mll 9M12

Red-orange Yellow Yellow Yellow Yellow Yellow Purple Red Pink Orange Purple Orange-brown (remaining bands)

TABLE VI Fraction

Band color

9M12A 9M12B 9M12C 9M12D 9M12E 9M12F 9M12G 9M12H 9M12I

Orange Orange Purple Yellow Brown Yellow Yellow Brown Brown

Isolation of Spinochrome from E. calamari8.-A 700-g batch of spines was processed for crude pigments as described above. Essentially the same array of pigments that was found in E. diadema was displayed. The only significant differences that were noted are (1) 2,5-dihydroxy-3-ethylbenzoquinonecould not be detected and (2) fraction 1 exhibited seven pigments instead of one for E. diadema. Fraction 1was chromatographed on a thin-layer plate with benzene and separated into a red band (lA), three yellow bands (lB, IC, and l D ) , a red band ( l E ) , a yellow band (lF), and a redorange band (1G). Fraotion 1E proved to be identical with 2-hydroxy-3-acetylnaphthazarin(1). Fraction 1F was too small to crystallize, but its ultraviolet spectrum and Rr value were identical with those of 2-hydroxy-6-ethyljuglone (2). Fraction 1G crystallized from isooctane to give 0.25 mg of 2-hydroxy-6ethylnaphthazarin(3), mp 204-204.5'.

VOL.31

Conversion of 2-Hydroxy-6-acetylnaphthazarinto 2-Hydroxy-6ethylnaphthazarin .-Two milligrams of the natural pigment, 2hydroxy-6-acetylnaphthazarin, was carefully methylated with diazomethane and after purification by tlc the resulting 2methoxy-6-acetylnaphthazarin,red crystals from isooctane, mp 185-186' dec, was dissolved in 2 ml of methanol and ea. 1 mg of sodium borohydride was added (red color immediately discharged). The yellow solution was acidified with hydrochloric acid (red color slowly returns) and the reduction product was extracted with benzene and separated from unreduced material by thin layer chromatography. The 2-methoxy-6-ethylnaphthazarin, which was not isolated, was hydrolyzed to 2-hydroxy6-ethylnaphthazarin by a 2-min boiling in 1 ml of ethanol-12 N hydrochloric acid (1:l). The product was identified as 2hydroxy-6-ethylnaphthazarinby comparison of its ultraviolet spectrum and Rt value (tlc) with those of an authentic sample. On a thin layer plate of acid-treated, deactivated silica gel, 2hydroxy-6-ethylnaphthazarinmoves faster than the 7-ethyl isomer in carbon tetrachloride. Synthesis of 2,5-Dihydroxy-3-ethylbenzoquinone.-The quinone was prepared using a modified procedure of Jones and ShonleZ4for 2,5-dihydroxybenzoquinone. A stirred mixture of 40 mg of ethylhydroquinone in 1 ml of 60% aqueous sodium hydroxide was trettted with 0.5 ml of 30% hydrogen peroxide. The temperature rose to 45' and was maintained a t 45-50' for 2 hr during which time the mixture became a thick paste. It was poured onto ice and acidified with hydrochloric acid. The product was extracted from the resulting yellow solution with ether and purified by vacuum sublimation. After crystallization from benzene 13 mg of 2,5-dihydroxy-3-ethylbenzoquinone(27%) was obtained as orange prisms, subliming a t 130-145' without melting. The synthetic quinone was identical in every respect (nmr, infrared and ultraviolet spectra, and Rt values is two systems) with the natural pigment.

Acknowledgment.-We should like to thank Mr. Lester C. Zukeran of the Hawaii Institute of Marine Biology for collecting the animals and Dr. D. Becher and Professor C. Djerassi for determining the mass spectra. The authors are grateful to Miss Glory Barayuga for invaluable technical assistance. This investigation was supported by a PHS Grant GM 10413 from the Institute of General Medical Sciences, Public Health Service. (24) R. G. Jonesand H. A. Shonle, J . Am. Chem. Soc., 67, 1034 (1945).

Mass Spectrometry in Structural and Stereochemical Problems.' CXI. The Mass Spectrometric Fragmentation of Substituted Naphthoquinones and Its Application to Structural Elucidation of Echinoderm Pigments' DIETER BECHER,* CARLDJERASSI, RICHARD E. MOORE,HARJITSINGH, AND PAUL J. SCHEUER Contribution from the Departments of Chemistry, Stanford University, Stanford, California 94306,and University of Hawaii, Honolulu, Hawaii 96888 Received March 16, 1966 The ma8s spectra of a series of acetyl, and higher substituted naphthoquinones are presented. The generalizations set out by Beynon and William and by Williams and co-workers have been expanded by considering the fragmentation patterns characteristic for acetylnaphthoquinones. The acetyl function surpasses the hydroxy and methoxy function in its ability to direct fragmentation upon electron h p a c t . Its location in the quinoid or benzenoid moiety of the molecule and in the vicinity of various other substituents yields different spectra, which can be used successfully for structural elucidation of unknown pigments of this group.

Naphthoquinone pigments occur widely in nature among higher plants and microorganisms, yet in the animal kingdom this class of compounds has been encountered only in echinoderms. Even within this (1) Paper CX: A. M. Duffield, 9. D. Sample, and C. Djerassi, Chem. Commun., 193 (1966). (2) Supported in part by Grants No. GM-11309 and GM 10413 (PHS). (3) Recipient of a NATO Postdoctoral Fellowship.

phylum only the echinoids (sea urchins) have been the prime producer of these substances. There they occur as structural pigments (spinochromes) and in the eggs, ovaries, and perivisceral fluid of the animals (echinochromes). The first research in this field dates back to 1885,4but it was more than 50 years later when a pure, (4) C. A. MacMunn, Quart. J . Microacop. Sci., 46, 488 (1885).